Fracture and lineament patterns across the Midcontinent indicate post-Cretaceous reactivation of 1 basement-involved faults 2 3

8 Reactivation of pre-existing weaknesses in the upper crust can be documented using surface features, and 9 has occurred throughout time and space, particularly in regions where the basement material dates from 10 the Precambrian and has undergone successive deformation events. This study aims to use surface 11 features such as fracture patterns to document evidence of such reactivation in the Paleozoic and 12 Cenozoic of Nebraska and Kansas (units separated by an unconformity in the study area). The most 13 prominent basement features in southeast Nebraska and northeast Kansas are oriented NE-SW, likely 14 related to the midcontinent rift, and oriented NW-SE, likely related to fabrics from the Central Plains 15 Orogen. These features are well defined in the potential fields data. Fracture patterns in the study area 16 show an E-W oriented trend, as well as clearly discernable NE-SW and subsidiary N-S and NW-SE trends. 17 The E-W trend is interpreted to be related to far-field stresses from Laramide and Ancestral Rocky 18 Mountain orogenic events, whilst the NE-SW trend is interpreted to be related to subtle reactivation on 19 the Mid-continent rift and related faults, observed in basement data. These movements produced 20 stresses of sufficient magnitude to produce extensional fractures in the overlying rock units, but not 21 sufficient to generate shear. Similarly, the ~N-S and NW-SE fracture trends are taken as evidence of subtle 22 reactivation on the Nemaha Uplift and Central Plains Orogen systems, generating fractures but not shear 23 movement. This contribution therefore provides a convincing case-study of the value of fracture 24 orientations (that is, surface morphodynamics) in discerning buried tectonic trends and subtle 25 reactivation thereon. 26 27 Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 8 April 2018 doi:10.20944/preprints201804.0092.v1 © 2018 by the author(s). Distributed under a Creative Commons CC BY license. Peer-reviewed version available at Geosciences 2018, 8, 215; doi:10.3390/geosciences8060215


1.INTRODUCTION
In many landscapes, surface evidence of deformation can provide a tantalizing glimpse of the structures beneath the surface.Landscape maturity can be used to determine the order in which folds amplified and to discern the presence of subtle folding structures that cannot be observed in other ways (e.g.Obaid & Allen, 2017).Landslide occurrence and the relief of the local topography can also be employed to understand the pattern of active faults in a region (e.g.Osmundsen et al., 2009).Cinque et al. (1993) use geomorphological domains in addition to subsurface datasets to infer the geodynamic evolution of the Southern Apennines.These three diverse studies highlight the importance of geomorphologic data in subsurface geology.In this contribution, we use the shape of the land surface, as gleaned from satellite data and field-derived fracture patterns, to understand the reactivation history of basement-involved faults in SE Nebraska and NE Kansas.
Reactivation of pre-existing weaknesses in crustal material is documented throughout time and space, particularly in regions where the basement material dates from the Precambrian and has undergone successive episodes of deformation.The concept of tectonic inheritance -that the pre-existing structures and zones of weakness in a system govern the development of subsequent structures -is frequently used to explain large-scale variations in the geometry of orogenic belts, or the locations of rift margins in supercontinent cycles (Butler et al., 1997, Thomas 2004, Audet & Burgmann, 2011, Huerta & Harry, 2012).
In addition, this concept can be used to explain the appearance of folds and faults or uplifted zones in regions that are far from the collision zone.The effect of basement geometry on the development of an orogenic arc has been demonstrated by Macedo & Marshak (1999) using analog models.Large-scale finite-element modeling also demonstrates that a pre-existing structural system can exert significant control on subsequent deformation styles (Huerta & Harry, 2012).Studies of specific structures (e.g.Gates & Costa, 1998;Molliex et al., 2010;Said et al., 2011;McMechan, 2012;Burberry, 2015) indicate that subsequent structures, facies changes and economic deposits may be affected by the motion of the preexisting fault or faults.
Failed rifts and ancient suture zones can provide weak zones suitable for reactivation in successive deformation events.The amount of reactivated movement seen on any given fault is a function of the lithospheric strength, number of faults within the sequence, and obliquity between the fault orientation and compression direction (Dewey 1989, Williams et al. 1989, Del Ventisette et al. 2006).Pressure and temperature of the deforming system may also affect reactivation potential (Ranalli, 2000).Major reactivated faults have a mix of effects, such as compartmentalizing the overlying thrust belt structure, nucleating overlying folds, as well as affecting the basement geometry (Butler et al. 2006).Depending upon the compression direction, initially normal faults may be reactivated as strike-slip or as reverse/thrust faults in subsequent deformation if the conditions are correct, that is, the incident stress is within a suitable angle relative to the original fault (Handin, 1969).Some studies indicate that an oblique subsequent imposed stress is more favorable for fault reactivation than an imposed stress normal to the structure (Sibson, 1985).Faults may be reactivated numerous times if the region is affected by multiple phases of deformation, assuming that the above conditions are met in each phase.The effect of an imposed stress on a pre-existing weakness in a specific area can be tested using analog models, where the initial conditions can be chosen and scaled to represent a real-world scenario (McClay, 1995).This methodology has been applied successfully to regions including the Italian Alps and the Zagros Belt (Bahroudi & Koyi, 2003;Viola et al., 2004).
This study aims to demonstrate that basement structures can reactivate and propagate through a thick sequence of cover rocks to influence surface fracture patterns.The study area chosen is the southeastern section of Nebraska and the adjacent northwestern region of Kansas (Figure 1).This area was chosen because there are undisputed structures within the basement namely the southern extent of the Midcontinent Rift System and the northern extent of the Nemaha uplift.Through a comparison of lineaments derived from potential fields processing and analysis, remotely sensed lineaments and surface fracture patterns, we demonstrate that reactivation of basement faults is able to explain the distribution of surface lineaments (that is, probable faults and fractures) in the study area.Previous studies have documented fracture patterns across parts of the present study area (e.g.Neff, 1949;Ward, 1968) and noted the relationship of the surface lineaments to presumed basement features (e.g.Baehr, 1954;Smith et al., 1974;White, 1990).Although providing valuable data, these previous studies cover only a small portion of the present focus area, and lack the understanding of modern rock mechanics in explaining their findings (e.g.Neff, 1949;Nelson, 1952).This study serves to emphasize the influence of basement feature reactivation on surface geomorphology in the framework of modern rock mechanics, and conversely, the importance of surface features in deciphering the history of a region.

GEOLOGIC SETTING
The earliest crust in southeast Nebraska and northern Kansas is considered to date from the Central Plains Orogen (CPO trend on Figure 2) around 1.9-1.7 Ga (Carlson, 1995).This crust is crystalline basement, in the area of interest, this is quartzite with granite intrusions (e.g.Jewett & Merriam, 1959;Anderson & Wells, 1968).The study area is dissected by two major structures, the NE-SW trending, 1.1 Ga Mid-Continent Rift System (MRS) and the NNE-SSW trending 300 Ma Nemaha Uplift (NU), as can be seen on contour maps of the top-Precambrian unconformity (e.g.Jewett & Merriam, 1959;Condra & Reed, 1959) and the present Figure 2. The MRS is considered an aulacogen, and has a counterpart failed rift arm to the east.The Nebraska segment of the MRS parallels fabrics within the earlier Central Plains Orogen (Carlson, 1995).In the southern part of the study area, within eastern Kansas, the MRS is cut by a series of NW-SE trending dextral strike-slip faults, on which there have been historical earthquakes (Baars, 1992;Carlson & Treves, 2005).These faults appear to form an accommodation zone between the Nebraskan and Kansan segments of the MRS (Figure 2) and may be related to the boundary of the Central Plains Province or the suture between this province and the Penokean province which is located to the southeast of the MRS (Carlson, 1995(Carlson, , 1997;;Berendsen, 1997).The structures are also aligned along the same trend as the Missouri Gravity Low which underlies the accommodation zone as well as additional NW-SE trending lineaments interpreted from gravity and magnetic data (Atekwana, 1996).More detailed maps correlate specific trends to accretion of numerous terranes during the Central Plains Orogen (Carlson, 2007;Whitmeyer and Karlstrom 2007).
The effects of local tectonics during the Infracambrian and the Paleozoic are observed from stratigraphic relationships, thickness changes and multiple unconformities within the Cambrian-Permian succession, particularly in the present study area (Figure 3; Anderson & Wells, 1968;Carlson, 1997Carlson, , 1998;;Burberry et al., 2015).The Cambrian-Devonian succession is chiefly composed of dolomite, with some clastic units; these clastic materials may indicate periods of land exposure and influx of eroded material to an area otherwise covered by a shallow sea (Jewett & Merriam, 1959;Scotese & Golonka, 1992).Subsequently, extensive Mississippian carbonate rocks were deposited in a shallow sea, across large swathes of the midcontinent (Goebel, 1968).The boundary between the Mississippian and the Pennsylvanian rocks is marked by a regional unconformity, showing widespread uplift of the region in the late Mississippian and early Pennsylvanian (e.g.Scotese & Golonka, 1992).
Inversion via compression of MRS system structures is documented from calcite strain analysis in the Lake Superior Region, ranging in age from Grenvillian-age transpressive strain to Permian age contraction (Craddock et al., 1997).Hauser (1996) also documents Grenville-age inversion on the MRS and crosssections after Eardley (1962) andJewett &Merriam (1959) show significant deformation in the Cambrian-Mississippian section, overlain by a base Pennsylvanian unconformity.A major tectonic event, making the NU emergent, is documented in the Ordovician (Anderson & Wells, 1968;Berendsen and Speczik, 1991;Berendsen et al., 1992).The Kansas segment of the MRS may have been reactivated in the latter part of the Paleozoic, forming the Abilene, Voshell, Barneston, and Nemaha anticlines along the eastern edge of the MRS (Mendenhall, 1958;Carlson, 1995).Rift segments are also considered to be reactivated in the formation of the Union fault and related faults of the study area (Carlson, 1997;Burberry et al., 2015).Many workers (e.g.Baars, 1992;Wilson & Berendsen, 1998) consider the NU to be a reactivated fault that originated as part of the MRS, affected by the segmented nature of the rift and the difference in rift orientation between Nebraska and Kansas (Carlson, 1997).The Nemaha uplift is imaged by a COCORP line as a 40 km wide uplift, bounded to the east by the near-vertical Humboldt Fault (Brown et al., 1983;Serpa et al., 1989;Gay, 1999).The Humboldt fault is thought to be a multiply reactivated structure originally related to the MRS, currently showing a transpressive sense of motion (Moore, 1926;Gay, 1999).
The predominant bedrock in the study area is of Pennsylvanian and Permian age.The Pennsylvanian strata are represented by the mixed clastic and shallow-water carbonate facies of the Desmoinesian, Missourian and Virgilian series (Figure 3;Heckel, 2008).The lowermost Pennsylvanian Atokan series is not expected to be present in the study area (Condra & Reed, 1959).The Virgilian Series of the Upper Pennsylvanian (Figure 3) is the oldest material that crops out in the study area (Figure 1; Heckel 2013).The Virgilian series is characterized by 11 of the short, intense highstand interglacials (Heckel, 2008) that make up the Middle and Upper Pennsylvanian.Each of these cyclothems consists of a black-gray shale unit, overlain by a marine limestone consisting of regressive then shoaling upwards facies, overlain by paleosol (Heckel, 2008).Despite the regular presence of shale members in the cyclothems, the net shale is sufficiently low that the shale layers do not act as detachment units (Condra & Reed, 1959;Jewett & Merriam, 1959).
Both the Wolfcampian and Leonardian Series of the Lower Permian crop out in the study area (Figure 1).
The Wolfcampian series continues the cyclothem succession from the Upper Pennsylvanian (Moore, 1964) and indicates that the Midcontinent Sea (Heckel 2008) still covered the area.The Leonardian Series is made up from dominantly clastic units, including thin anhydrite layers (Condra & Reed, 1959;Jewett & Merriam, 1959;West et al., 2010).This study samples sites from the carbonate units of the Wolfcampian cyclothems, and does not sample the clastic series.
Thickness changes in the Pennsylvanian and Permian sediments around the NU (e.g.Burberry et al., 2015) indicate active tectonics during this period, probably as a result of the Ancestral Rocky Mountain Orogen (ARM: Moore, 1926;Kluth & Coney, 1981;Gay, 1999;Joeckel et al., 2007).This orogeny is a result of the combined stresses on the intraplate region from the contemporaneous Antler orogeny to the west and the Ouachita-Marathon orogeny (i.e. the docking of Gondwana) to the south-east.In support of the reactivation on the NU during the Pennsylvanian-Permian, Merriam & Forster (2002) and Underwood and Poison (1988) document a series of earthquake proxy locations, including intraformational faulting and contorted bedding in the Pennsylvanian and Permian section.
Following significant uplift and exposure during the Triassic and Jurassic, the region formed part of the shoreline of the Cretaceous Western Interior Seaway during the Late Cretaceous.This period is marked by the deposition of the Dakota Group (Condra & Reed, 1959;Jewett & Merriam, 1959).No Triassic or Jurassic age rocks are recorded in the study area (Figure 1).The Dakota Group includes alternating reddish, clay-rich paleosols and clean, cross-bedded sandstone units, some of which were sampled at sites NF 1, NF 2 and KF 1. Overlying the Dakota group is the Graneros Shale and the Greenhorn Formation, representing periods of incursion of the Seaway and deposition of shales, followed by the thinly-bedded carbonate and shale layers of the Greenhorn Formation (Condra & Reed, 1959).The Greenhorn Formation is sampled at location KF 2 and is the youngest Formation sampled in this study.
After the ARM-related reactivation pulses, the NU may have been uplifted for a final time during the Laramide orogeny (Burberry et al., 2015) as the Laramide stress field is favorably oriented for transpressional movement with respect to many basement structures in the midcontinent (Tikoff & Maxson, 2001;Ohlmacher & Berendsen, 2005).Historic earthquakes have occurred in the study area, notably related to the southern boundary of the MRS and the eastern boundary of the NU (Ohlmacher & Berendsen, 2005), indicating that stress directions are still favorable for fault reactivation.However, overall, the region is considered to have been essentially tectonically quiescent since the Eocene, despite minor seismicity along structures such as the Nemaha anticline (Steeples et al., 1979).
Figure 3 indicates that there is no thick ductile unit in the study area, when the low ratio of shales to rigid units such as limestone and sandstone is taken into account (for a more detailed stratigraphy, refer to Condra & Reed (1959) and Jewett & Merriam (1959).Despite the units labeled as "interbedded x and shale", the shale horizons have not acted as major decollement surfaces.The only documented fault activity is intraformational faulting (Neff, 1949;Merriam & Forster, 2002) and these authors do not describe large-scale detachment-style behavior.Thus, it is reasonable to assume (a) that the Paleozoic sedimentary succession is behaving as a rigid beam and (b) that reactivation on deep-seated faults could have influenced the deformation of the entire rigid beam.

METHODS
The first phase of this study was carried out using a combination of remote sensing and field data collection.The remote dataset used was a series of Landsat Thematic Mapper images, obtained from NASA and processed under the MrSID algorithm (Tucker et al., 2004).This algorithm combines Band 7 (mid-infrared light) as red, Band 4 (near-infrared light) as green and Band 2 (visible green light) as blue.
This produces a false color image in which bare rock surfaces are colored in shades of pink and brown (clastic units are darker than carbonate units) and vegetation appears in shades of green.Water typically appears dark blue-black (Figure 4).These images have a ground resolution of 28.5 m.Given that the study area is heavily vegetated, anomalies in the drainage network and variations in the vegetation patterns were used to interpret surface lineaments.Care was taken to avoid man-made irrigation systems (locally referred to as "waterways" by soil conservation personnel) and vegetation changes related to crop patterns; typically distinguishable from natural variation by scale and geometry.Man-made irrigation systems are generally smoothly curving, continuous with, or at right angles to, the pattern of terraces in the field, and confined to individual fields or sections.Vegetation changes related to crop patterns are typically oriented N-S or E-W in the study area and are typically perfectly straight lines.In contrast, natural drainage systems display patterns of tributaries, run for long distances and are related in some fashion to the underlying geology.Unexpected drainage geometries, such as straight segments with no evidence of man-made alteration, or sharp bends in a system were marked and used to guide the interpretation of lineaments.
Fracture data was collected at thirteen field sites, seven in SE Nebraska and six in NE Kansas, forming two transects across the study area.Transect 1 is made up from sites NF 1-NF 7 and is oriented E-W.Transect 2 is made up from sites KF 1-KF 6 and is oriented NE-SW.These data were collected in part because of limited outcrop availability and in part to have transects running broadly perpendicular to the anticipated trends of the main basement features.Bedding orientation data as well as fracture strike and dip was measured at each location.Outcrops were frequently road cuts, thus some bias is introduced into the data by the orientation of the road cut -for example, on an E-W oriented road cut, any fractures oriented E-W will be under-represented.This cannot be avoided, but was carefully noted for use in analysis.At each site, 25 fracture orientations were measured.Care was taken to avoid radial fractures caused by dynamiting the outcrop in the road-construction process.
To gain a sense on the orientation of basement structures, we mapped lineaments using spatial analysis of gravity and magnetic grids (Bankey et al., 2002, Kucks, 1999).The magnetic anomaly grid was extracted from the North American Magnetic Anomaly Map database, comprising a compilation from numerous vintage airborne surveys that were leveled to a consistent elevation of 305 m above the terrain.In our study area, the flight line spacing varied from 0.5 -8 km (Bankey et al. 2002).The total magnetic intensity data were reduced to magnetic pole in order to remove the skewness of magnetic signals due to nonverticality of the ambient field (inclination of 69.65°, a declination of 9.46°).
The gravity anomaly grid from Kucks, 1999 was used for the study.This dataset compiles the numerous land gravity measurements by the US Geological Survey, corrected for the rocks above sea level with an assumed density of 2.67 g/cc (Bouguer gravity anomaly).In order to remove the long-wavelength crustal signal from the gravity data, we generated the regional trend via upward continuation of observed gravity grid to a 100 km elevation.By removing this regional trend from the observed Bouguer gravity, we obtained the map of the residual Bouguer gravity anomalies, which represent the gravitational signal due to lateral density distribution in the subsurface rocks.
To further highlight the subsurface structures, we applied tilt derivative filters (Verduzo et al., 2004) to both the reduced to pole magnetic map and the residual Bouguer gravity grid.This procedure highlights the zones of the subtle changes in the potential fields in both vertical and horizontal directions.In gravity, an additional 5-km wide low-pass filter was necessary in order to remove some original gridding artifacts.
The resultant filtered gravity and magnetic fields (Figure 5  4. RESULTS: SURFACE LINEAMENT/FRACTURE DATA 8,255 surface lineaments were mapped in the study area using Landsat images as described above, and the orientation of each lineament was calculated as a bearing.Only a subset of the main area is shown for ease of viewing (Figure 6).Lineaments visible in Figure 5 are predominantly oriented NE-SW.When all lineaments from the study area are displayed on a rose diagram (Figure 7), three prominent orientations can be noted (arrowed in Figure 7) -at 005, 055 and 085.About 8% of the data follows the 055 orientation, forming the most significant orientation in the study area.Secondary peaks at 115 and 145 can also be noted in Figure 7 (not arrowed) although these orientations are only just above the apparent background or uniform distribution.illustrate the data collected at each field site, after minor rotations were made to the dataset due to bedding dip. Figure 8 shows pole figures from the E-W oriented transect, NF 1-NF 7 and Figure 9 shows rose diagrams for the same data.Site NF 1 is in the Dakota Formation, at a highly weathered outcrop including numerous iron concretions south of Fairbury, NE.The sandstone at this outcrop is friable, but has fractured cleanly with a fracture spacing of about 1 m.There are three main fracture orientations, 057, 088 and 110, shown clearly by the clusters on the pole figure.No abutting relationships were discernable at this outcrop.Site NF2 is a blocky, systematically fractured outcrop of the Dakota Sandstone, south of NF 1.There are two prominent clusters on the pole figure, one of which can be separated into three orientations.These four orientations are 016, 040, 053 and 125.NF 3 is an outcrop of the Permian Wolfcampian series at a waterfall on private land.A thin, fractured limestone layer overlies a reddish shale unit.The pole figure appears to be a combination of both of the previous two, with fracture orientations prominently at 002, 031, 047, 090 and 122.NF 4 is a thinly bedded, highly fractured outcrop with chert nodules, along Plum Road between Liberty, NE and Wymore, NE.The units come from the Permian Wolfcampian Series (Figure 1).Prominent fracture orientations in this outcrop are at 029, 072 and 117.Site NF 5 is in the Virgilian Series of the Upper Pennsylvanian (Figure 1, Figure 3).The outcrop is a road cut along Highway 4, west of Table Rock, NE.A 1.5 ft thick layer of fossiliferous limestone is present, overlying thinly bedded limestone and shale.Bedding planes contain fossil hash, including spiriferids, other brachiopods, fenestrate bryozoan fragments, and crinoid ossicles.Prominent fracture orientations are 004, 026, 049, 068, 096 and 115.Site NF 6 is in the Virgilian Series of the Upper Pennsylvanian (Figure 1, Figure 3).The outcrop consists of a ridge of black shale overlain by a blocky recrystallized limestone containing crinoid fragments, thereby placing the outcrop close to the highstand of one of the Upper Pennsylvanian cyclothems.The outcrop is a quarry face within a succession of limestone and shale that is 18 ft thick and forms part of a fault scarp (Andy Keller, Martin Marietta Plant Manager, pers.comm.).The prominent fracture orientations are 034, 090, 105, 135 and 173.Lastly, site NF 7 is located close to a substation west of Humboldt, NE.The outcrop is a 1-ft thick layer of carbonate mudstone which is systematically fractured into blocks.The outcrop is considered to overly the Humboldt Fault (R.M. Joeckel, pers. comm.).Prominent fracture orientations are 019, 062, 073 and 173.Table 1 summarizes the pronounced fracture orientations at each outcrop, with the ETM (remote sensing) dataset included for comparison.The "mean" row in the table records the average orientation of each set of fractures, based on information in the table.The bolded numbers represent fracture sets appearing in both the field and remote datasets.The 001, 059, 090 and 110 orientations are found in both the ETM dataset and in 5 or more of the field sites.The 148 orientation is only found in one field site, but is prominent in the ETM dataset.Closer inspection of the table reveals that there is a subtle spatial trend appearing in the dataset, namely that orientations 137 and 163 only appear in the southern part of the study area (southern part of transect NF and the KF transect).However, neither of these trends are apparent in the ETM dataset.
Considering the data by age is also instructive.Figure 12 shows rose diagrams and pole figures for the Cretaceous, Permian and Pennsylvanian units, respectively, and Table 2 shows a summary of major fracture orientations.Note that the number of fractures in each plot is significantly different as there are more outcrops in the Permian units than in either Cretaceous or Pennsylvanian.Considering the dataset by age shows that the 001 orientation appears in all datasets, as does the 090.Some important orientations, e.g. the 031 and 072 orientations only appear in the older units.Similarly, some orientations, such as the 124 and 163 orientations, are only prominent in the Permian and younger units.Finally, in this presentation of the data, orientations such as the 020, 046 and 059 only appear in the youngest (that is, the Cretaceous age) units.

RESULTS: BASEMENT LINEAMENTS FROM POTENTIAL FIELDS DATA
In the study area, 47 lineaments were mapped through a joint analysis of the filtered magnetic and gravity maps (Figure 13).Many of the lineaments correlate well with previously mapped basement faults (Figure 2).The interpreted lineaments show a strong NE-SW trend that mirrors the orientation of the MRS.This trend can clearly be seen when the plotted on a rose diagram (Figure 14).The majority of the lineaments (36%) are oriented at 025 and 045.In addition to the MCR parallel orientations, a strong NW-SE trend at 125 is also apparent with a smaller spike in the 145 direction.

COMPARISON OF SURFACE AND BASEMENT DATASETS
The basement lineaments in Figure 14 show the most tightly clustered orientations, when compared to Figures 7, 9, 11 and 12 (all other rose diagrams).The two prominent basement trends neatly mirror the CPO and MRS system faults known from other work (Figure 2).The CPO trend is not as strongly observed in the study area as is the MRS trend.Table 3 shows that all basement orientations appear in the surface dataset, and the overall NE-SW orientation appears in all three datasets, at either ~045 or ~055.This indicates that the surface datasets, both fracture and ETM-derived, are picking up the basement features but with varying amounts of dispersion.Figure 15 shows that there are areas (e.g.arrowed, marked S) where the ETM dataset is picking up the same orientation as the basement feature, and that there are areas (e.g. the region marked R) where the ETM dataset appears to be showing Riedel shears to the main basement lineament, as well as a similar orientation.Comparison of Figure 15 to Figures 8 and 9 (NF 3) shows that the main NE-SW ETM trend is on-trend with a basement lineament, but is a more minor trend in the measured fractures at this site.Overall, the surface fracture trends and the basement lineaments line up nicely, but the ETM dataset introduces a significant amount of noise into the system.This correspondence of lineament and fracture orientations is best explained by repeated reactivation on the basement faults, as detailed in the "Geological Setting" section, throughout the Pennsylvanian, Permian (both ARM-related) and lower Cretaceous.A Cretaceous phase of reactivation is less well documented than the other ARM-related reactivation, and could be due to far-field stresses from the Laramide Orogeny.

DISCUSSION
As noted above, the basement lineaments (Figures 13,14) mirror the trends of Precambrian age faults related to the MRS and the CPO.Overall, the correspondence of surface lineaments and fractures to basement data indicates reactivation of basement structures in an important influencing factor in the development of the surface features in Nebraska and Kansas.Our results enhance the work of Neff (1949), Nelson (1952), Baehr (1954), Ward (1968), Smith et al. (1974) and White (1990) by extending the dataset of fractures and satellite-derived lineaments, and documenting the relationships with basement features over a wider area.We have further shown that the strong presence of the MRS-related basement orientation in the Cretaceous age datasets (Figures 12,14) provides strong evidence for reactivation of the MRS trends in the mid Cretaceous.
The surface and ETM datasets are much more dispersed than the basement dataset.We first need to consider that the depth to the basement is significant in the study area, leading to attenuation of the gravity and magnetic signature of buried structures.As a result, only dominant subsurface features will be represented in the basement analysis, compared to a much wider range of features represented at the surface.One possibility for dispersion in surface datasets, alluded to above, is the generation of secondary fracture orientations during reactivation as strike slip faults.This can be seen in the development of fractures in Riedel shear orientations (Figure 15).Another possibility is that flexure and uplift of the rigid beam of Phanerozoic sediments is distributing deformation across a wider area, perhaps by outer-arc extension of the thick beam above a vertically uplifting basement feature.Lastly, fractures are known to develop in unroofing, that is, uplift and erosion, situations, although these are typically small-scale and more randomly oriented.Lineament analysis on this scale and at this resolution is unlikely to pick up many unroofing structures, but some of the noise in the field datasets (Figures 8-12) may be attributed to this.An additional factor that must be considered in the evaluation of Figure 12 is that sandstone layers typically fracture in a much blockier, systematic fashion than the underlying limestones.Limestone containing numerous fossil fragments appears to fracture with more dispersion around the key orientations than a sandstone layer of a similar thickness, subjected to the same stresses (see e.g. a, b) show multiple lineaments associated with subsurface structures.The joint interpretation of the lineaments in both potential fields resulted in a map of potential faults in the basement rock.

Figures
Figures 10 and 11 show pole figures and rose diagrams for field sites located along the NW-SE oriented transect in Kansas (KF 1-KF 6 respectively).Site KF 1 is an E-W oriented road cut exposing about 3 m of the Dakota Group.In this location, the Dakota Formation is manifest as a deep orange, cross-bedded sandstone, with evidence of re-precipitated or concentrated iron in bedding planes and cross-bedding.Prominent fracture sets are oriented003, 022, 060, 084, 117 and 139.Site KF 2 is an outcrop of the Greenhorn Formation near Cuba, KS; a thinly bedded limestone and shale road cut containing inoceramids and oriented E-W.At this location, prominent fracture orientations are 008, 037, 111, 136 and 165.Site KF 3 is an outcrop of cherty, thickly-bedded limestone within the Wolfcampian Series of the Permian.The road cut is oriented N-S and is systematically fractured.Prominent fracture orientations at this outcrop are 359, 029, 068, 090, 110 and 137.Site KF 4 is also within the Permian Wolfcampian Series, and is a thinly-bedded, cherty limestone.The site is located close to the Fancy Creek State Park area of Tuttle Creek Lake.Prominent fracture orientations are 357, 029, 049, 072, 106, and 148.This is the only outcrop that shows the 148 orientation.Site KF 5 is within the Permian Wolfcampian Series and is a weathered outcrop close to the Tuttle Creek Spillway and the Spillway Fault System.Prominent fracture orientations here are 003, 021, 060, 072, 137 and 162.Lastly, site KF 6 is located south of Manhattan, KS, near the Konza Prairie Kansas Valley lookout point.The road cut is oriented N-S and is made up of interbedded limestone and shale from the Permian Wolfcampian Series.Prominent fracture orientations are 359, 020, 045, 078, 100 and 161.

Figures and Captions Figure 1 Figure 1 :Figure 2 Figure 2 :
Figures and CaptionsFigure1

Figure 8 :
Figure 8: Contoured poles to planes for fractures at each site on the E-W transect.Sites are arranged in order from W to E. The number of fractures in each figure is shown at the bottom left.

Table 1 :
Summary of fracture orientations for all field sites, organized by transect.The ETM data is given 388

Table 2 :
Summary of main fracture orientations in field sites organized by age.The mean and ETM data 392

Table 3 :
Summary of key fracture orienrataions from each dataset.The mean orientations from surface fractures, orientations from ETM analysis and orientations from potential fields analysis are shown.